Vm-dev November 2023

vm-dev@lists.squeakfoundation.org

17 participants
42 discussions

5.0-202311211431-64bit
by ken.dickey＠whidbey.com 27 Nov '23

27 Nov '23

A bit more info on RasPi4/Raspian/Cuis unwind test failures. Unable to get a fail using the stack interpreter (squeak.stack.spur) 20% fail (80% green) testing Process* unit tests with JIT (squeak.cog.spur). Each time the fail is in a different test: 'ProcessTerminateUnwindNonLocalReturn>>#test3DResume' 'ProcessTerminateBug>>#testTerminationDuringNestedUnwind7' . 'ProcessTerminateBug>>#testTerminationDuringNestedUnwind4' . 'ProcessTerminateUnwindNonLocalReturn>>#test5BTerminate' 'ProcessTerminateUnwindNonLocalReturn>>#test3BResume' 'ProcessTest>>#testValueEnsured' . 'ProcessTerminateBug>>#testTerminationDuringNestedUnwind3' . All tests done on same Cuis image (latest Cuis with `Feature require: 'BasicImageTests'.`) Repeated tests vs restart of same image +test do not seem to differ. E.g. 1st unit test run is sometimes green, sometimes yellow. FYI, -KenD

1 0

[SqueakJS] Smalltalk Debugger support
by Vanessa Freudenberg 27 Nov '23

27 Nov '23

In recent images, SqueakJS often does not allow proper debugging, most of the buttons are disabled, e.g. just executing "self halt" in a workspace: [image: image.png] self interruptedProcessIsReady => false interruptedProcess isSuspended => false myList isNil => false If, instead of sending #halt I do a "debug it", it works fine. In that case, "myList" is nil and interruptedProcessIsReady answers true. Does anyone have a clue what the difference might be? Vanessa

2 1

Current Squeak on VM Candidate
by ken.dickey＠whidbey.com 27 Nov '23

27 Nov '23

Squeak6.0-22104 on Raspberry Pi 4 under Raspian Linux. Rebuilt squeak.cog.spur build.assert from sources: opensmalltalk-vm-202311211431.tar.gz I am still seeing sqAllocateMemorySegmentOfSizeAboveAllocatedSizeInto mmap: Cannot allocate memory when I run the unit tests before updating, but not after. See attached screen captures for unit test details before and after Update Squeak. FYI, -KenD

1 0

VM Maker: VMMaker-dtl.440.mcz
by commits＠source.squeak.org 24 Nov '23

24 Nov '23

David T. Lewis uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker-dtl.440.mcz ==================== Summary ==================== Name: VMMaker-dtl.440 Author: dtl Time: 24 November 2023, 3:41:35.092 pm UUID: d347df2d-ee8f-4564-8178-f628d7327704 Ancestors: VMMaker-dtl.439 Fix undefined C behavior issue for primitiveMultiply and bytecodePrimMultiply.. Overflow must be tested in advance, otherwise aggressive C optimizers lead to e.g. 16r30000000 * 4 ==> -1073741824. Code adopted from oscog for primitiveMultiply, also updated similarly for bytecodePrimMultiply. =============== Diff against VMMaker-dtl.439 =============== Item was changed: ----- Method: ContextInterpreter>>bytecodePrimMultiply (in category 'common selector sends') ----- bytecodePrimMultiply + | rcvr arg result overflow | - | rcvr arg result | rcvr := self internalStackValue: 1. arg := self internalStackValue: 0. (self areIntegers: rcvr and: arg) ifTrue: [rcvr := objectMemory integerValueOf: rcvr. arg := objectMemory integerValueOf: arg. + overflow := rcvr > 0 + ifTrue: [arg > 0 + ifTrue: [rcvr > (objectMemory maxSmallInteger / arg)] + ifFalse: [arg < (objectMemory minSmallInteger / rcvr)]] + ifFalse: [arg > 0 + ifTrue: [rcvr < (objectMemory minSmallInteger / arg)] + ifFalse: [(rcvr < 0) and: [arg < (objectMemory maxSmallInteger / rcvr)]]]. + overflow + ifFalse: [result := rcvr * arg. + (arg = 0 or: [(result // arg) = rcvr and: [objectMemory isIntegerValue: result]]) + ifTrue: [self internalPop: 2 thenPush: (objectMemory integerObjectOf: result). + ^ self fetchNextBytecode "success"]]] - result := rcvr * arg. - (arg = 0 or: [(result // arg) = rcvr and: [objectMemory isIntegerValue: result]]) - ifTrue: [self internalPop: 2 thenPush: (objectMemory integerObjectOf: result). - ^ self fetchNextBytecode "success"]] ifFalse: [self initPrimCall. self externalizeIPandSP. self primitiveFloatMultiply: rcvr byArg: arg. self internalizeIPandSP. self successful ifTrue: [^ self fetchNextBytecode "success"]]. messageSelector := self specialSelector: 8. argumentCount := 1. self normalSend. ! Item was changed: ----- Method: InterpreterPrimitives>>primitiveMultiply (in category 'arithmetic integer primitives') ----- primitiveMultiply + | integerRcvr integerArg integerResult overflow | - | integerRcvr integerArg integerResult | integerRcvr := self stackIntegerValue: 1. integerArg := self stackIntegerValue: 0. self successful ifTrue: + [overflow := integerRcvr > 0 + ifTrue: [integerArg > 0 + ifTrue: [integerRcvr > (objectMemory maxSmallInteger / integerArg)] + ifFalse: [integerArg < (objectMemory minSmallInteger / integerRcvr)]] + ifFalse: [integerArg > 0 + ifTrue: [integerRcvr < (objectMemory minSmallInteger / integerArg)] + ifFalse: [(integerRcvr < 0) and: [integerArg < (objectMemory maxSmallInteger / integerRcvr)]]]. + overflow + ifTrue: [self primitiveFail] + ifFalse: + [integerResult := integerRcvr * integerArg. + self pop: 2 thenPush: (objectMemory integerObjectOf: integerResult)]]! - [integerResult := integerRcvr * integerArg. - "check for C overflow by seeing if computation is reversible" - ((integerArg = 0) or: [(integerResult // integerArg) = integerRcvr]) - ifTrue: [self pop2AndPushIntegerIfOK: integerResult] - ifFalse: [self primitiveFail]]! Item was added: + ----- Method: ObjectMemory>>maxSmallInteger (in category 'interpreter access') ----- + maxSmallInteger + ^1073741823! Item was added: + ----- Method: ObjectMemory>>minSmallInteger (in category 'interpreter access') ----- + minSmallInteger + ^-1073741824! Item was changed: ----- Method: VMMaker class>>versionString (in category 'version testing') ----- versionString "VMMaker versionString" + ^'4.20.6'! - ^'4.20.5'!

1 0

Squeak Alpha Unit tests on VM Candidate 202311211431
by ken.dickey＠whidbey.com 24 Nov '23

24 Nov '23

Greetings, I am trying out the Squeak6.0alpha-21503-64bit-202112201228-Linux-ARMv8 image on the OpenSmalltalk VM release candidate and am getting some odd results. Perhaps you can help me out. Running all unit tests in the image as downloaded gives quite a few errors and failures (see attached; RasPi4/Raspian/arm64/squeak.cog.spur). Trying to "Update Squeak" from the menu causes the RasPi to hang! Seems to kill the OS as I can't even ping the machine when this happens. I tried the same on an arm64 Pine64 laptop and get the same result running unit tests on the downloaded image. On this system (Manjaro Linux), I _am_ able to update the Squeak image, but get a hunk image running MCMczInstallerTest>>#testInstallFromStream . Interestingly, I get a hang in this same test on Risc64/Debian/VisionFive2 under squeak.stack.spur. MCMczInstallerTest>>#testInstallFromStream with the diagnostic sqAlloateMemorySegmentOfSizeAboveAllocatedSizeInto mmap: Cannot allocate memory I am building debug and assert VMs, but meanwhile feel free to help me to enlightenment. Cheers, -KenD ================================NOTES================================= RasPi4/Raspian Linux Squeak6.0alpha-21503-64bit-202112201228-Linux-ARMv8/shared '5283 run in 0:00:06:08.081, 5042 passes, 44 expected failures, 40 failures, 157 errors, 0 unexpected passes' ================================ Raspian:RasPi4:~ >>> squeak --version 5.0-202311211431 Tue 21 Nov 2023 04:06:06 PM PST gcc 10 [Production Spur 64-bit aarch64 VM] CoInterpreter VMMaker.oscog-eem.3339 uuid: b1c834e2-5de7-464c-aeb5-4f2a7b8812d6 Nov 21 2023 StackToRegisterMappingCogit VMMaker.oscog-eem.3339 uuid: b1c834e2-5de7-464c-aeb5-4f2a7b8812d6 Nov 21 2023 VM: 202311211431 pi@Raspian4:OpenSmalltalk/opensmalltalk-vm Date: Tue Nov 21 15:31:40 2023 CommitHash: 5b1e8d926 Plugins: 202311211431 pi@Raspian4:OpenSmalltalk/opensmalltalk-vm Linux Raspian4 6.1.21-v8+ #1642 SMP PREEMPT Mon Apr 3 17:24:16 BST 2023 aarch64 GNU/Linux plugin path: /usr/local/bin/../lib/squeak/5.0-202311211431-64bit [default: /usr/local/lib/squeak/5.0-202311211431-64bit/] Raspian:RasPi4:~ >>> uname -a Linux Raspian4 6.1.21-v8+ #1642 SMP PREEMPT Mon Apr 3 17:24:16 BST 2023 aarch64 GNU/Linux Raspian:RasPi4:~ >>> ======================

1 0

[SqueakJS] Release 1.1.2
by Vanessa Freudenberg 24 Nov '23

24 Nov '23

Hi all, I don't regularly do these announcements here, but this one fixed a bug that was reported 9 years ago <https://github.com/codefrau/SqueakJS/issues/23>, and I finally fixed it with the help of Agustin Martinez, who narrowed down the issue to the exact BitBlt call that produced the wrong result. The issue was in the color mapping code: function rgbMapPixelflags(sourcePixel, mapperFlags) { var val; val = SHIFT((sourcePixel & cmMaskTable[0]), cmShiftTable[0]); val = val | (SHIFT((sourcePixel & cmMaskTable[1]), cmShiftTable[1])); val = val | (SHIFT((sourcePixel & cmMaskTable[2]), cmShiftTable[2])); return val | (SHIFT((sourcePixel & cmMaskTable[3]), cmShiftTable[3])); } where SHIFT is function SHIFT(a, b) { return b < 0 ? (b < -31 ? 0 : a >>> (0-b) ) : (b > 31 ? 0 : a << b); } This is used for swizzling color components in the PNGReadWriter, but rarely anywhere else so it didn't affect the system otherwise, and wrongly led us to believe that it's a PNG-specific issue. The problem wasn't even in this code but here: cmShiftTable = loadColorMapShiftOrMaskFrom(interpreterProxy.fetchPointerofObject(0, cmOop)); with function loadColorMapShiftOrMaskFrom(mapOop) { if (mapOop.isNil) { return null; } if (typeof mapOop === "number") { interpreterProxy.primitiveFail(); return null; } if (!(interpreterProxy.isWords(mapOop) && (SIZEOF(mapOop) === 4))) { interpreterProxy.primitiveFail(); return null; } return mapOop.words; } This was auto-generated from the Slang BitBlt code back in 2014. Do you see the problem? This is the fix <https://github.com/codefrau/SqueakJS/commit/cefa684b57c3984113b82936f448c54…>. (Instead I could have fixed SHIFT() but this seemed simpler) Anyways, I thought it was interesting :) The release also adds object pinning <https://github.com/codefrau/SqueakJS/commit/63ad01221e1b96799215b54db129bd1…> and other things Cuis relied on, as well as supporting keyboard input in ancient Scratch images <https://github.com/codefrau/SqueakJS/issues/153>. Vanessa

2 1

[OpenSmalltalk/opensmalltalk-vm] BitBlt rgbMul is broken for depth 1 and not correctly rounded above depth 4 (Issue #651)
by Nicolas Cellier 24 Nov '23

24 Nov '23

In depth 1, the resulting bits will always be 0. It's not a big problem because rgbMul is just a bitAnd operation at this depth. So a quick workaround would be to detect the case in BitBltSimulation destDepth = 1 ifTrue: [^self bitAnd: sourceWord with: destinationWord]. That would also accelerate the Bit BLock Transfer operation, so it's a good hack. But there is more. What we want is multiply ratios in interval [0,1]. dstRatio * srcRatio Our implementation is scaled ratio (scaled by `1 << nBits - 1`): src := (srcRatio * scale) rounded. dst := (dstRatio * scale) rounded. So what we want is: ((dst/scale) * (src/scale) * scale) rounded that is: (dst*src / (1<<nBits-1)) rounded Unfortunately, that's the other grief with the current implementation used for rounding: (dst+1)*(src+1) - 1 >> nBits It only equals correctly rounded operation for depths 2 and 4. For rounding we might use: (((dst/scale) * (src/scale) + 0.5) * scale) truncated. that is expressed with truncated division: dst*src + (scale+1//2) // scale So here is a nicer formulation for doing the job at any depth (including 5bits rgb channels for 16 bits depth) with correctly rounded division: aux := src * dst + (1 << (nBits - 1)). "add mid-scale for rounding" result := aux << (nBits - 1) + aux << (nBits -1). "divide by scale" This is because instead of dividing by scale, we can multiply by shifted inverse (sort of double precision), then shift right. (2 to: 32) allSatisfy: [:nBits | (1 << (nBits * 2) / (1 << nBits - 1)) rounded = (1 << nBits + 1)]. Multiplying by this inverse is easy and cheap: x * (1 << nBits + 1) = (x << nBits + x). And then applying the right shift `>> (2 * nBits)` is equivalent to: x >> nBits + x >> nBits. We must first add 0.5 (scaled), that is `src * dst + (1 << (nBits -1))` - our formulation of aux, and we're done. We verify: { (0 to: 1<<20-1) allSatisfy: [:i | (1<<9+i)>>10+ (1<<9+i)>>10 = (i/1023) rounded]. (0 to: 1<<18-1) allSatisfy: [:i | (1<<8+i)>>9+ (1<<8+i)>>9 = (i/511) rounded]. (0 to: 1<<16-1) allSatisfy: [:i | (1<<7+i)>>8+ (1<<7+i)>>8 = (i/255) rounded]. (0 to: 1<<14-1) allSatisfy: [:i | (1<<6+i)>>7+ (1<<6+i)>>7 = (i/127) rounded]. (0 to: 1<<12-1) allSatisfy: [:i | (1<<5+i)>>6+ (1<<5+i)>>6 = (i/63) rounded]. (0 to: 1<<10-1) allSatisfy: [:i | (1<<4+i)>>5+ (1<<4+i)>>5 = (i/31) rounded]. (0 to: 1<<8-1) allSatisfy: [:i | (1<<3+i)>>4+ (1<<3+i)>>4 = (i/15) rounded]. (0 to: 1<<6-1) allSatisfy: [:i | (1<<2+i)>>3+ (1<<2+i)>>3 = (i/7) rounded]. (0 to: 1<<4-1) allSatisfy: [:i | (1<<1+i)>>2+ (1<<1+i)>>2 = (i/3) rounded]. } allSatisfy: #yourself. The nice thing is that above down-scaling operation can be multiplexed. Suppose that we have p groups of 2*nBits `M` holding square-scale multiplication of each channel concatenated in a double-Word-Mul. doubleWordMul = Mp .... M5 M3 M1 Note we arrange to have odd channels in low word, and even channels in high word. We first form a `groupMask` on a word with (p+1)/2 groups of nBits alternating all one `i` and all zero `o`, `oioi...ioi`. channelMask := 1 << nBits - 1. groupMask := 0. 0 to: wordBits // (2 * nBits) do: [:i | groupMask = groupMask << (2 * nBits) + channelMask]. Where wordBits is the number of bits in a word (usually we want to operate on 32 bits words in BitBlt). We form the `doubleGroupMask` on a double-word with p groups of 2*nBits `oi`: doubleGroupMask := groupMask >> nBits. doubleGroupMask := doubleGroupMask << wordBits + groupMask. And we perform the division by scale: doubleWordMul := (doubleWordMul >> nBits bitAnd: doubleGroupMask) + doubleWord >> nBits bitAnd: doubleGroupMask. At this stage we obtain a double word containing scaled multiplicands interleaved with groups of nBits zeros: o mp ... o m3 o m1 Now the final result can be obtained by shifting back: doubleWordMul >> (wordBits - nBits) + (doubleWordMul bitAnd: groupMask) The only problem remaining is how to obtain the squared-scale multiplicands. It would be easy to form the alternate even-odd channels for each src and dst operands: doubleWordSrc := src >> nBits bitAnd: groupMask. doubleWordSrc := doubleWordSrc << wordBits + (src bitAnd: groupMask). doubleWordDst := dst >> nBits bitAnd: groupMask. doubleWordDst := doubleWordDst << wordBits + (dst bitAnd: groupMask). we now get `o sp ... o s3 o s1` and `0 dp ... o d3 o d1`, but we would now need a SIMD integer multiplication operating on groups of 2*nBits in parallel... We don't have that, at least in portable C code. So we still have to emulate it with a loop. half := 1 << (nBits - 1). shift := 0. doubleWordMul := 0 0 to: nChannels - 1 do: [:i | doubleWordMul := doubleWordMul + (((doubleWordSrc >> shift bitAnd: channelMask) * (doubleWordSrc >> shift bitAnd: channelMask) + half) << shift). shift := shift + nBits + nBits]. We know that each operation cannot overflow on upper neighbour group of 2*nBits, because the maximum value is: (1<<nBits-1) squared + (1 << (nBits-1)) = 1 << (2*nBits) - (2*(1<<nBits)) + (1 << (nBits-1)) - 1 < (1 << (2*nBits) - 1) It remains the odd case of 16 bits depth, which has 3 groups of 5 bits and a leading zero. I believe that above algorithm works without splitting in two half-words... To be tested. We have gathered the pieces for a correctly rounded almost-multiplexed rgbMul. Somehow have our cake and eat it too. -- Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/651 You are receiving this because you are subscribed to this thread. Message ID: <OpenSmalltalk/opensmalltalk-vm/issues/651(a)github.com>

2 4

Cuis Basic Image Tests on VM Candidate build 5.0-202311211431
by ken.dickey＠whidbey.com 24 Nov '23

24 Nov '23

VM built and run on same processor in Linux ===================================== Riscv64 squeak.stack.spur - BaseImageTests 1936 run 1934 Passed, 2 failed, 0 errors BitBltTest>>#testAllAlphasRgbMul ProcessorTest>>testGrabProcessor ===================================== Aarch64 squeak.stack.spur 1936 run 1934 Passed, 1 failed, 0 errors BitBltTest>>#testAllAlphasRgbMul ===================================== Aarch64 squeak.cog.spur 1936 run 1934 Passed, 1 failed, 0 errors BitBltTest>>#testAllAlphasRgbMul =====================================

3 5

VM Maker: CogTools-LM.94.mcz
by commits＠source.squeak.org 23 Nov '23

23 Nov '23

Leon Matthes uploaded a new version of CogTools to project VM Maker: http://source.squeak.org/VMMaker/CogTools-LM.94.mcz ==================== Summary ==================== Name: CogTools-LM.94 Author: LM Time: 23 November 2023, 5:13:14.611051 pm UUID: bf8819e7-2069-49db-a7d9-593bfad324c2 Ancestors: CogTools-eem.93 Filter Linux symbols on an image level, not in the primitive. Follow-up for VMMaker.threaded-LM.3348 This ensures that even if the primitive leaves an entry of the Array empty, the Manager just filters it out. =============== Diff against CogTools-eem.93 =============== Item was changed: ----- Method: VMProfilerLinuxSymbolsManager>>initializeMost (in category 'initialize-release') ----- initializeMost | shortNames | initialized := false. maxAddressMask := (2 raisedToInteger: 32) - 1. modulesByName := Dictionary new. symbolsByModule := Dictionary new. shortNames := Set new. tempDir := self class tempDirectory. CompatibilityClass ensureExistenceOfDirectory: tempDir. modules := self primitiveExecutableModules. modules := (1 to: modules size by: 2) collect: [:i| | fileName shortName counter longName | fileName := modules at: i. + "The linux-gate.so.1 library may result in an empty entry." + fileName ifNotNil: [ + (fileName beginsWith: '/dgagent') ifTrue: + [fileName := fileName allButFirst: 8]. + shortName := CompatibilityClass nameOfFile: fileName in: tempDir. + counter := 0. + [shortNames includes: shortName] whileTrue: + [counter := counter + 1. + shortName := (CompatibilityClass nameOfFile: fileName in: tempDir), counter printString]. + shortNames add: shortName. + longName := (modules at: i + 1) + ifNil: [fileName] + ifNotNil: + [:symlink| + symlink first = $/ + ifTrue: [symlink] + ifFalse: [( CompatibilityClass parentPathOfFile: fileName ), '/', symlink]]. + "some files are off limits (e.g. /dgagent/lib/preload.so)" + (CompatibilityClass exists: longName) ifTrue: + [(modulesByName + at: longName + put: VMPExecutableModuleSymbol new) + name: longName; + shortName: shortName]]]. - (fileName beginsWith: '/dgagent') ifTrue: - [fileName := fileName allButFirst: 8]. - shortName := CompatibilityClass nameOfFile: fileName in: tempDir. - counter := 0. - [shortNames includes: shortName] whileTrue: - [counter := counter + 1. - shortName := (CompatibilityClass nameOfFile: fileName in: tempDir), counter printString]. - shortNames add: shortName. - longName := (modules at: i + 1) - ifNil: [fileName] - ifNotNil: - [:symlink| - symlink first = $/ - ifTrue: [symlink] - ifFalse: [( CompatibilityClass parentPathOfFile: fileName ), '/', symlink]]. - "some files are off limits (e.g. /dgagent/lib/preload.so)" - (CompatibilityClass exists: longName) ifTrue: - [(modulesByName - at: longName - put: VMPExecutableModuleSymbol new) - name: longName; - shortName: shortName]]. "The primitive always answers the VM info in the first entry." vmModule := modules first. "now filter out the files we can't read..." modules := modules select: [:m| m notNil and: [modulesByName includesKey: m name]]!

1 0

VM Maker: VMMaker.threaded-LM.3348.mcz
by commits＠source.squeak.org 23 Nov '23

23 Nov '23

Leon Matthes uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.threaded-LM.3348.mcz ==================== Summary ==================== Name: VMMaker.threaded-LM.3348 Author: LM Time: 23 November 2023, 5:07:58.598833 pm UUID: c652b1df-5728-4cc0-89ef-c735a6134916 Ancestors: VMMaker.threaded-LM.3347 Fix VMProfileLinuxSupportPlugin>>#primitiveExectuableModules 64-bit linux handles the linux-gate.so.1 differently which lead to memory-corruption due to the primitive expecting to filter out a single item. =============== Diff against VMMaker.threaded-LM.3347 =============== Item was changed: ----- Method: VMProfileLinuxSupportPlugin>>primitiveExecutableModules (in category 'primitives') ----- primitiveExecutableModules "Answer an Array of pairs of strings for executable modules (the VM executable and loaded libraries). The first element in each pair is the filename of the module. The second element is either nil or the symlink's target, if the filename is a symlink." <export: true> | resultObj | numModules := 0. self dl_iterate_phdr: #countnummodules _: 0. resultObj := interpreterProxy instantiateClass: interpreterProxy classArray + indexableSize: numModules * 2. - indexableSize: numModules - 1 * 2. "skip the fake linux-gate.so.1" resultObj = 0 ifTrue: [^interpreterProxy primitiveFailFor: PrimErrNoMemory]. interpreterProxy pushRemappableOop: resultObj. primErr := numModules := 0. self dl_iterate_phdr: #reapmodulesymlinks _: 0. resultObj := interpreterProxy popRemappableOop. primErr ~= 0 ifTrue: [^interpreterProxy primitiveFailFor: primErr]. ^interpreterProxy methodReturnValue: resultObj! Item was changed: ----- Method: VMProfileLinuxSupportPlugin>>reap:module:symlinks: (in category 'iteration callbacks') ----- reap: info module: size symlinks: ignored "like reap:module:names:, but follows symlinks" <var: 'info' type: #'struct dl_phdr_info *'> <var: 'size' type: #'size_t'> <var: 'ignored' type: #'void *'> <returnTypeC: #int> | elfModuleName len moduleNameObj symLinkBuf | <var: 'elfModuleName' type: #'const char *'> <var: 'symLinkBuf' declareC: 'char symLinkBuf[PATH_MAX]'> elfModuleName := numModules > 0 ifTrue: [info dlpi_name] ifFalse: [self getAttributeString: 0]. (elfModuleName isNil or: [(len := self strlen: elfModuleName) = 0]) ifTrue: + [^0]. "skip the fake linux-gate.so.1 --- NOTE: On 64-bit linux this is linux-vdso.so.1 + and doesn't seem to appear as a zero-length/nullptr string!! + It also doesn't seem to hurt anything to leave it in the list." - [^0]. "skip the fake linux-gate.so.1" moduleNameObj := interpreterProxy instantiateClass: interpreterProxy classString indexableSize: len. moduleNameObj = 0 ifTrue: [primErr := PrimErrNoMemory. ^1]. "stop iteration" self strncpy: (interpreterProxy arrayValueOf: moduleNameObj) _: elfModuleName _: len. "(char *)strncpy()" interpreterProxy storePointer: numModules ofObject: interpreterProxy topRemappableOop withValue: moduleNameObj. "now dereference the symlink, if it exists" self str: symLinkBuf cpy: elfModuleName. (len := self read: elfModuleName li: symLinkBuf nk: #'PATH_MAX') > 0 ifTrue: [moduleNameObj := interpreterProxy instantiateClass: interpreterProxy classString indexableSize: len. moduleNameObj = 0 ifTrue: [primErr := PrimErrNoMemory. ^1]. "stop iteration" self strncpy: (interpreterProxy arrayValueOf: moduleNameObj) _: symLinkBuf _: len. "(char *)strncpy()" interpreterProxy storePointer: numModules + 1 ofObject: interpreterProxy topRemappableOop withValue: moduleNameObj] ifFalse: [interpreterProxy storePointer: numModules + 1 ofObject: interpreterProxy topRemappableOop withValue: interpreterProxy nilObject]. numModules := numModules + 2. ^0!

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev November 2023